Thanks, Georg, for the comments. I'll take these into account and re-work the code a bit. If you can point me towards the resultfile flag in your playground, I'll have a look at that, too, though perhaps not immediately, as there are more pressing bugs to fix before 1.5.rc1.
Richard Georg Baum wrote: > Am Freitag, 18. Mai 2007 21:14 schrieb José Matos: > >> On Friday 18 May 2007 7:42:56 pm Richard Heck wrote: >> > > >>> The attached patch addresses these bugs, all of which have to do with >>> export, viewing, and the like---ultimately, with the conversion >>> routines. The patch introduces a new conversion flag, usetempdir, which >>> forces the converter to do all its work in a temporary directory >>> > created > >>> under the lyx_tmpbuf directory. On export, this directory is copied to >>> the directory in which the LyX file is contained. So, if you open >>> /path/file.lyx and File>Export>Format, you will get a directory >>> /path/file.format.conversion/ in which all the files generated by the >>> converter will be found (e.g. the mess that htlatex generates ends up >>> > in > >>> /path/file.html.conversion/). >>> > > A flag for the name of that directory does already exist: resultdir. Please > use that instead of /path/file.html.conversion/. > > >>> If you're just viewing, however, you get >>> $LYXTMPDIR/lyx_tmpbufN/format/, and of course that will get deleted >>> > when > >>> LyX exits (gracefully). >>> >>> As a result, the originaldir flag does not need to be used by any of >>> > our > >>> converters, and it may be redundant, though there's no harm leaving it, >>> so far as I can see. >>> > > IMHO it should be removed. It does not fit into the "copy everything to > temp dir" strategy, and complicates the code. In the cases when it was > formerly needed (a converter needed to access files in the original dir) a > specialized mover can be used nowadays (as is done e.g. for xfig). > > >>> As you will note, doing this involved adding a new signature to the >>> Converters::convert() routine. I have not changed any of the other >>> > calls > >>> to this: these are in Importer.cpp, insets/InsetGraphics.cpp, and >>> insets/ExternalSupport.cpp. At least some of these will need modifying >>> before the patch is applied, probably all of them, as I suppose >>> usetempdir could be set for importers and graphics converters and such. >>> > > IMHO there should only be one interface, and I don't like the new one: If > the resulting file name is determined by the converter in some cases, then > why not do this always? AFAIK the callers of the converter determine the > result from the old name and the format extension anyway, so this should > be easy to change. > Then there is some error checking missing: If the caller can't deal with a > converted directory (e.g. for preview) then it should check that the > result is indeed a file and act appropriately if that is not the case. > > BTW there is another problem in this area that I solved in my playground > branch: Some converters don't accept a commandline argument to set the > resulting file name, but determine it using their own rules. They only > work if these rules match the naming given by LyX, but will fail if a user > changes e.g. the extension of a format. I solved this by a new resultfile > flag. > > >>> There are a few other clean-up type things to be done still, too, >>> including fixing indentation, but I haven't done that yet as it would >>> make the diff hard to read. >>> >> This area looks a mess and I have never quite understood it >> > completely. :-) > > This area is a big mess indeed. AFAIK the originaldir flag never really > worked, and HTML export has been broken for all but the most simple cases > for ages. > > I don't want to look at the patch in detail, but the general direction is > sound. > > > Georg > -- ================================================================== Richard G Heck, Jr Professor of Philosophy Brown University http://frege.brown.edu/heck/ ================================================================== Get my public key from http://sks.keyserver.penguin.de Hash: 0x1DE91F1E66FFBDEC Learn how to sign your email using Thunderbird and GnuPG at: http://dudu.dyn.2-h.org/nist/gpg-enigmail-howto
